RISK OF STROKE AND DEATH IN
ELDERLY DEMENTED PATIENTS
TREATED WITH ATYPICAL
ANTIPSYCHOTICS
Rajesh R. Tampi, MD, MS, FAPA
Associate Clinical Professor of Psychiatry
Yale University School of Medicine,
rajesh.tampi@yale.edu
DISCLOSURES

I have no financial relationships to disclose.

All of the medications discussed today are “offlabel”, as no agents have been approved by the
FDA for the treatment of dementia-related
psychosis or for the treatment of behavioral
and psychological symptoms of dementia
(BPSD).
TIMELINE OF EVENTS

2002: Health regulatory authorities in Canada and
Switzerland were the first to raise concerns about
the association of risperidone with CVAEs in clinical
trials of elderly dementia patients.

Spring of 2003: FDA followed suit and published
similar warnings for risperidone.

January of 2004: FDA added a warning of
increased CVAEs with olanzapine.

2004: Eli Lilly releases similar warnings and made
changes to prescribing information for olanzapine.

March 2004: UK’s Committee on Safety of
Medicines (CSM) advised prescribers that
risperidone and olanzapine should not be used for
the treatment of BPSD, because of “clear evidence
of an increased risk of stroke”.

February 2005: FDA added a warning of increased
CVAEs with aripiprazole.

April 2005: FDA Public Health Advisory: Deaths
with Antipsychotics in Elderly Patients with
Behavioral Disturbances

The Food and Drug Administration has determined that the treatment of
behavioral disorders in elderly patients with dementia with atypical (second
generation) antipsychotic medications is associated with increased mortality.

17 placebo controlled trials in elderly demented patients with behavioral
disturbance, involving olanzapine, aripiprazole, risperidone, or quetiapine,
enrolling a total of 5106 patients, were reviewed.

15 showed numerical increase in mortality (1.6-1.7 fold) in the drug-treated
group compared to the placebo-treated patients.

Examination of the specific causes of these deaths revealed that most were
either due to heart related events (e.g., heart failure, sudden death) or infections
(mostly pneumonia).

The warning also extended to other atypical antipsychotic medications including
clozapine, ziprasidone and combination olanzapine, fluoxetine.

The FDA also added a similar warning to older antipsychotic medications
because the limited data available suggested a similar increase in mortality for
these drugs.
THE NEW YORK TIMES APRIL 11, 2005
RISK OF CEREBROVASCULAR ADVERSE
EVENTS AND DEATH IN ELDERLY PATIENTS
WITH DEMENTIA WHEN TREATED WITH
ANTIPSYCHOTIC MEDICATIONS: A
LITERATURE REVIEW OF EVIDENCE.
MITTAL V, KURUP L, WILLIAMSON D, MURALEE S,
TAMPI RR. AM J ALZHEIMERS DIS OTHER DEMEN.
2011 FEB;26(1):10-28

The following search terms were used ‘stroke’, ‘cerebrovascular events’,
‘cerebrovascular accidents’, ‘death’, ‘dementia’, ‘behavioral symptoms’, ‘behavioral
disturbance’, ‘BPSD’, ‘antipsychotics’ and ‘psychotropics’. The terms were searched
within the following databases - Cochrane Library, Medline, EMBASE, and PsycINFO for
English articles in humans from January 1, 1990- August 31, 2010.

Of all the articles identified from the search, articles were selected that commented
on CVAE and death with the use of antipsychotic treatment in dementia patients.

These articles consisted of randomized controlled trials, pooled analysis, metaanalyses, systematic reviews, population based studies, prospective studies,
retrospective studies, reviews and case reports.

In addition, a manual search of bibliographies and the FDA website was done. The
results are described for CVAEs followed by death.

We organized the studies in the results section into subsections from the highest to
the lowest level of evidence. Placebo controlled trials were given first priority followed
by pooled analysis, meta-analysis and systematic reviews. Population based studies,
retrospective analyses and other studies were considered lower on the level of
evidence. We further arranged the studies in each of the subsections based on the
date of publication to maintain a chronological order.

Results (CVAEs)

In our review of the literature, we found 22 studies that evaluated the risk of CVAEs.

Of these 22 studies, only two were prospective studies. The majority of the studies
(14) were population based studies, retrospective analysis or other studies.

The available data indicates that the risk of CVAEs is higher in the drug treated group
(1.3-2.0 times).

Although preliminary, existing data for atypical versus typical antipsychotics indicate
that the risk of CVAEs is similar in both groups.

No one drug has been found to be safer than the other in terms of the CVAEs.

A higher than median doses of a drug, older age, a diagnosis of dementia especially
vascular dementia and comorbid atrial fibrillation have been noted as risk factors for
CVAEs.

It appears that the time frame for which the risk of CVAEs remains elevated is about
20 months.

Results (Death)

For the risk of death of death, we found a total of 14 studies which addressed this
risk.

Of these 14 studies, only three were placebo controlled trials.

Preliminary data indicates that risk of death with atypical and typical antipsychotics is
greater than when compared to the placebo group or the group that did not use these
medications.

The risk is about 1.2-1.6 times higher in the drug treated group.

Existing data for atypical versus typical antipsychotics indicate that the risk of death is
similar in both groups.

No one drug has been found to be safer than the other in terms of the death.

Older age, male gender, severe dementia and functional impairment are associated
with a higher risk of death.

The risk remains elevated from 30 days to 2 years.

Mechanisms:

The development of orthostatic hypotension, most likely as a result of antagonism at
alpha-adrenergic receptors.

In an individual with cerebrovascular insufficiency or atherosclerosis, they might
experience a CVAE as a consequence of hypotension aggravating the deficit in cerebral
perfusion.

Also, tachycardia induced by alpha-adrenergic antagonism might cause a decrease in
cerebral perfusion as a result of a rate-induced decrease in diastolic filling and cardiac
output or tachycardia might dislodge a thrombus in the patient with atrial fibrillation.

Antipsychotics have a variable degree of effect on prolactin levels by their actions on
dopamine transmission involved in the tuberoinfundibular tract. Theoretically,
hyperprolactinemia can accelerate atherosclerosis and increase the risk of TIAs and
strokes.

It is possible that sedation or obtundation might cause antipsychotic treated patient to
develop dehydration and hemoconcentration, leading on to the development of a
thrombus. Antipsychotic medications have also been associated with an increased risk of
venous thromboembolism. These thrombi could be potentially dislodged into the arterial
circulation, causing TIAs and strokes especially in patients with atrial fibrillation.

The mechanisms by which antipsychotic medications may contribute to death
remains to be established.

Most deaths are thought to be due to cardiovascular events (mostly
arrhythmias) and infections (pneumonia for the great majority).

Antipsychotic use has been associated with a lengthening of QTc interval on
electrocardiograms.

Recently, atypical medications, such as ziprasidone, quetiapine, risperidone,
and olanzapine, have been linked to QTc prolongation, although the highest
estimate of risk has been documented for thioridazine.

An increased risk of ischemic cerebrovascular events has been also linked to
the use of atypical antipsychotics in patients with dementia.

It must also be kept in mind that older patients with dementia are at higher
risk for cerebrovascular events and death per say and their comorbidities
along with the use of these medications only adds to this risk.
Do atypical antipsychotics cause stroke?
Herrmann N, Lanctôt KL. CNS Drugs. 2005;19(2):91-103.
 11 studies
 48 out of 2187 (2.2%) drug-treated subjects experienced
CVAEs vs. 10 out of 1190 (0.8%) placebo treated.
 The combined relative risk was 2.7 (95% CI, 1.4 to 5.3).
 Numerically more risperidone-treated patients (33 of 1009
[3.3%]) experienced CVAEs compared with olanzapine-treated
patients (15 of 1178 [1.3%]).
 The weighted relative risk was statistically significant for
risperidone (3.2, 95% CI, 1.4 to 7.2), but not for olanzapine
(1.8, 95% CI, 0.5 to 6.3).
European Agency for the evaluation of Medicinal Products
http://www.emea.eu.int/humandocs/PDFs/EPAR/abilify/089304en
6.pdf
 In 3 placebo-controlled trials of aripiprazole in elderly AD
patients.
 Mean age: 84 years, range: 78-88 years.
 1.3% of aripiprazole-treated patients reported CVAEs vs. 0.6%
of placebo-treated patients.
 This difference was reported not to be statistically significant.
 In a fixed-dose trial, there was a significant dose response
relationship for CVAEs in patients treated with aripiprazole.
EFFICACY AND ADVERSE EFFECTS OF
ATYPICAL ANTIPSYCHOTICS FOR DEMENTIA:
META-ANALYSIS OF RANDOMIZED, PLACEBOCONTROLLED TRIALS.
Schneider LS, Dagerman K, and Insel PS.
Am J Geriatr Psychiatry. 2006 March;14:3; 191-210.
RESULTS

Cerebrovascular Adverse Events (CVAEs):
 There were 63/3327 (1.9%) events in the
medication group versus 16/1728 (0.9%) in
placebo group.
 The
OR by meta-analysis was 2.13, with CI of
1.20-3.75 and p=0.009.
 Of
note, significantly increased risk was found
with risperidone with OR=3.43 (3.1 vs. 1.0%
pooled).
RISK OF DEATH WITH ATYPICAL
ANTIPSYCHOTIC DRUG TREATMENT
FOR DEMENTIA: META-ANALYSIS OF
RANDOMIZED PLACEBO-CONTROLLED
TRIALS.
Schneider LS, Dagerman KS, Insel P.
JAMA. 2005 Oct 19;294(15):1934-43.
OBJECTIVE
 Assess
the evidence for increased allcause mortality from atypical
antipsychotic drug treatment for patients
with dementia.
RESULTS

15 trials (9 unpublished), generally 10 to 12
weeks in duration, including 16 contrasts of
AAPs with placebo met criteria (aripiprazole
[n = 3], olanzapine [n = 5], quetiapine
[n = 3], risperidone [n = 5]).

11 trials were performed in nursing homes
and 4 with outpatients.
RESULTS

3 with aripiprazole, 603 patients
2 in nursing homes
and 1 with outpatients (10-week duration)

5 with olanzapine, 1184 patients

5 with risperidone , 1175 patients

3 with quetiapine, 391 patients
2 in nursing homes,
3 with outpatients, and 1 with a risperidone comparison (6- to 26-week
durations)
including the
outpatient trial above with an olanzapine comparison, 4 in nursing
homes, 1 with a haloperidol comparison (8- to 12-week durations)
all nursing home trials,
1 with a haloperidol comparison and another with rivastigmine (10- to
26-week durations)
RESULTS

A total of 3353 patients were randomized to study drug and
1757 were randomized to placebo.

All but one of the trials were sponsored and conducted by
drug manufacturers.

87% had Alzheimer disease

Mean age 81.2 (7.8) years old

70% were women

Mean MMSE was 11.3
RESULTS
Medication
No of events in
the treatment
group
No of events in
the placebo group
Odds Ratio
(95% chance)
Aripiprazole
21/603
6/348
1.73 (0.70-4.30)
Olanzapine
31/1184
6/478
1.91 (0.79-4.59)
Quetiapine
21/391
7/246
1.67 (0.70-4.03)
Risperidone
45/1175
22/779
1.30 (0.76-2.23)
Overall
118/3353
(3.5%)
41/1851
(2.3%)
1.54 (1.06-2.23)
p=0.02
RESULTS

Incidence in exposed: 118/3353=3.5%

Incidence in non-exposed: 41/1851=2.3%

Relative Risk: (Incidence in exposed)/(Incidence in non-exposed):
3.5/2.3=1.52

Absolute Risk Increase: 3.5% -2.3%=1.2% (upper limits on CI of 4% to
5%)

Number Needed to Harm: 1/ARI = 1/0.012 = 83 with a 95% CI ranging
from 53 to 1000. Thus, implying that there may be 1 death due to
atypical drug use for every 83 patients treated over 10 to 12 weeks

Numbers Needed to Treat (NNT): 4 to 12 in meta-analyses
RESULTS

Likelihood of Harm versus Help (LHHH): NNH/NNT:
83/4 (21) to 83/12 (7)

Thus, for every 7 to 21 persons helped in these
trials, there possibly will be 1 death.
 If
NNH is 1000, then for every 83 to 250
persons helped, there will be 1 death.
 If
NNH is 53, then for every 4 to 13 persons
helped, there will be 1 death.
RESULTS

Attributable Risk Percent (ARP): ARI/(risk of
death in the exposed group) X 100%. (3.5%2.3%)/3.5% x 100 = 34%.

Therefore, almost 34% of the total risk of
death in elderly patients with dementia who
are exposed to AAPs is due to AAPs.
TAKE HOME POINTS
1.
Death occurred more often over the first 8 to 12 weeks of treatment among
patients randomized to an atypical antipsychotic than to placebo, 118 [3.5%]
vs 40 [2.3%]
2.
The OR by meta-analysis was 1.54 with a 95% CI of 1.06 to 2.23, p = 0.02
3.
The RR by a random-effects model was 1.65 with a 95% CI of 1.19 to 2.29
4.
Likelihood of Harm versus Help (LHHH) indicates that for every 7 to 21 persons
helped, there will possibly be 1 death
5.
Excess mortality was not due to any particular atypical antipsychotic but could
only be appreciated when this class of medications were examined as a whole
6.
Subgroup analysis did not reveal differences between patients of lower
cognitive function, psychosis of AD, or inpatients versus outpatients
RISK OF DEATH IN ELDERLY USERS
OF CONVENTIONAL VS. ATYPICAL
ANTIPSYCHOTIC MEDICATIONS.
Wang PS, Schneeweiss S, Avorn J, Fischer MA,
Mogun H, Solomon DH, Brookhart MA.
N Engl J Med. 2005 Dec 1;353(22):2335-41.
METHODOLOGY

Retrospective cohort study involving 22,890
patients 65 years of age or older who had drug
insurance benefits in Pennsylvania and who began
receiving a conventional or atypical antipsychotic
medication, for the first time, between 1994 and
2003

Analyses of mortality rates and Cox proportionalhazards models were used to compare the risk of
death within 180 days, less than 40 days, 40 to 79
days, and 80 to 180 days after the initiation of
therapy with an antipsychotic medication.

The data was controlled for potential confounding
variables.
RESULTS
 In the first 180 days of use:
 17.9%
of patients receiving conventional
antipsychotics died
 14.6%
of patients receiving atypical antipsychotics
died
 This
difference was statistically significant, with
p<0.001
 Only
cancer, congestive heart failure, and HIV
conferred greater adjusted risks of death
RESULTS
Relative Risk of Death within 180 Days After Beginning Therapy
with Conventional as Compared with Atypical Antipsychotic
Medications
Time Period
Relative Risk (RR)
95% Confidence Interval
<40 days
1.56
1.37 to 1.78
40-79 days
1.37
1.19 to 1.59
80-180 days
1.27
1.14 to 1.41
>180 days
1.37
1.27 to 1.49
RESULTS
RESULTS

Conventional antipsychotic medications were
associated with a significantly higher adjusted
risk of death than were atypical antipsychotic
medications at all time intervals studied and in
all subgroups defined, according to the
presence or absence of dementia or nursing
home residency.

On average for every 100 patients treated with
a conventional agent instead of an atypical
drug there would be 7 additional deaths.
RESULTS
 The
greatest increase in the adjusted
risk of death for conventional
antipsychotics as compared with
atypical antipsychotics occurred with
higher doses of conventional agents
and during the first 40 days after
initiation of therapy.
TAKE HOME POINTS
1.
Conventional antipsychotics are more likely than
atypical agents to increase the risk of death
among elderly persons, RR = 1.37.
2.
On average for every 100 patients treated with a
conventional agent instead of an atypical drug
there could be 7 additional deaths.
3.
The risk of death with conventional antipsychotics
is greatest within the first 40 days of use and with
higher doses.
ANTIPSYCHOTIC DRUG USE AND MORTALITY
IN OLDER ADULTS WITH DEMENTIA.
Gill SS, Bronskill SE, Normand SLT, et al.
Ann Intern Med. 2007; 146:775-786.
METHODOLOGY

Population-based, retrospective cohort study

Ontario, Canada from April 1997 to March 2003

New use of antipsychotics

Risk for all-cause mortality was determined at 30, 60, 120,
and 180 days.

Comparisons: atypical antipsychotics versus no
antipsychotic use, conventional versus atypical
antipsychotic use

Groups were also divided by place of residence: community,
long-term care facility
RESULTS

27,259 matched pairs were identified

Mean age was > 80 for all groups

In both the community and long-term care
cohorts:
 75.2%
of atypical antipsychotic users started
risperidone
 60.2% of conventional antipsychotic started
haloperidol
RESULTS
RESULTS
FINAL THOUGHTS
1.
Older adults with dementia who are exposed to atypical
antipsychotics have a small but significant increase in
cerebrovascular adverse events and mortality.
2.
Use of conventional antipsychotics seems to confer an equal
risk for CVAEs and death as with atypical antipsychotic use.
3.
Therefore, caution must be used when treating these patients
with this class of medications.
4.
Rabins and Lyketsos (2005) suggest an approach that limits
the use of these drugs to situations in which “there is an
identifiable risk of harm to the patient or others, when the
distress caused by symptoms is significant, or when alternate
therapies have failed and symptom relief would be
beneficial”.
8.
Cardiovascular and cerebrovascular risks should be carefully considered
and documented before initiating treatment with an atypical antipsychotic
in a patient with a previous history of stroke, transient ischemic attack, or
myocardial infarction.
9.
Consideration should also be given to other potential etiologies of the
problem behavior such as delirium, under treated or over treated medical
illness, environmental triggers, etc.
10.
Prescribe medications at the lowest effective dosages.
11.
If there is no response to the medication within 4-6 weeks, consider
discontinuation of the drug.
12.
If the patient does respond to the medication, continued treatment should
be contingent on close monitoring for side-effects and control of risk
factors.
13.
Treaters must reevaluate benefits and risks frequently and consider
discontinuation of treatment when appropriate.
THANK YOU!